DeepSeek V3 API

DeepSeek's DeepSeek V3, through an OpenAI-compatible API. Providers quote it on an open exchange, and each request goes to the offer with the lowest estimated charge that meets your rules.

Providers compete for each request. The exchange ranks the offers that meet your rules by estimated charge. It fixes your maximum charge before the first token, and you pay for the tokens that the model uses.
The OpenAI and Anthropic APIs. Use the OpenAI or Anthropic SDK that you have. Set the base URL and your key, and send deepseek-v3 as the model.
Your data rules. Send x-liquid-retention: none and only offers whose provider does not keep your content compete for the request.

Facts

Model name: deepseek-v3
Maker: DeepSeek
Context window: 163.8K tokens
Input price: $0.320 per million tokens
Output price: $0.891 per million tokens
Providers: 1 independent provider
Regions: us

Providers

Architect: us, $0.320 input and $0.891 output per million tokens, $0.320 cache read per million tokens, 163.8K tokens of context

Scores

HLE: 2.9, #264 of 273
GPQA: 55.7, #198 of 250
AA-LCR: 29.3, #203 of 237
SciCode: 35.8, #109 of 123
Terminal-Bench Hard: 6.8, #131 of 166
LiveCodeBench: 35.9, #75 of 130
AIME 2025: 26.0, #78 of 100
MMLU-Pro: 75.2, #68 of 133
τ²-Bench: 22.8, #148 of 169
IFBench: 34.8, #143 of 178

Scores: Artificial Analysis (https://artificialanalysis.ai/)

Request

curl https://router.inference.ai.exchange/v1/chat/completions \
  -H "Authorization: Bearer $LIQUID_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "deepseek-v3",
    "messages": [{"role":"user","content":"Say hello."}]
  }'

Questions

What is DeepSeek V3?: Served by intent-router: each job goes to the cheapest healthy upstream for this model.
How much does DeepSeek V3 cost?: Offers now ask $0.320 per million input tokens and $0.891 per million output tokens. A job pays for the tokens it uses. The exchange picks the cheapest eligible offer.
What is the context length of DeepSeek V3?: The largest offer takes 163,840 input tokens. A smaller offer can take less. The exchange skips an offer that cannot take your message.
What can DeepSeek V3 do?: Its providers list Reasoning effort, JSON object, JSON schema, Frequency penalty, Logit bias, Log probabilities, Min-p sampling, Presence penalty, Repetition penalty, Seed, Stop sequences, Top-k sampling, Top-p sampling, Text, Tools.
What inputs and outputs does DeepSeek V3 support?: Text → Text.
Which providers serve DeepSeek V3?: 1 independent provider quote it now, with 1 offer. The table above ranks them by the cost of a typical job.
How do I call DeepSeek V3?: Send "deepseek-v3" as the model in an OpenAI or Anthropic request. The API section links to setup steps for each client.
When was DeepSeek V3 listed?: The exchange first listed it on 10/8/2026. This is not its release date.